Tag Archive | Public Opinion

Are most people ideological?

Nice little argument about opinion from economists Bryan Caplan vs. Robert P. Murphy, Jason Weeden, and Robert Kurzban.  The root of the argument is Caplan’s somewhat blithe assertion that:

There are countless issues that people care about, from gun control and abortion to government spending and the environment…  If you know a person’s position on one, you can predict his view on the rest to a surprising degree.  In formal statistical terms, political opinions look one-dimensional.  They boil down to roughly one big opinion, plus noise.

Kurzban and Weeden respond:

This is just getting weird. You explicitly said of individual issues that “If you know a person’s position on one, you can predict his view on the rest to a surprising degree.” And we ran a test of that proposition, showing that it can be pretty weak stuff, depending on the issue pair. (I’ve got a blog post showing similar points here:http://www.pleeps.org/2014/06/30/if-being-routinely-liberal-or-conservative-is-a-human-universal-why-is-it-true-only-of-recent-college-educated-whites/)

Well…leaving aside the issues of tone and proof, they’re kind of both correct.   What Caplan is describing is generally thought of as “constraint”; highly-constrained people “know” what ideological positions go together.  Highly-constrained people look exactly like Caplan describe.  On the other hand, loosely constrained people are generally less informed and don’t necessarily have well-defined political opinions.  They also don’t know which opinions go together, and so generally have very low correlation between their opinions.  Most people are pretty loosely constrained, except for the highly educated.  This explains, incidentally, both why constraint is low in general and why it seems like only college-educated follow the pattern Caplan describes.

Finally, unlike Caplan I am happy to cite literature.  It’s from a book called “The American Voter”, published…in 1960.  Political scientists actually know what they’re doing, and economists could look at their work sometimes.  On a sad sidenote, one of the authors Philip Converse, just passed away.

On Good Bad Research

There’s a viral study going around – you might have run into it on Facebook – on porn consumption and marriage.  It went viral because duh, and its conclusion is that porn is destroying marriage.  Unfortunately, it’s bunk.  Jordan Weissman does an admirable job debunking the study; long story short, the authors use a terrible instrumental variable.  It’s a nice little look at the use and abuse of social science tactics, and how fancy math can be used to confidently frame complete bull.

This is bad research, and there’s a big market for bad research.  People love to be able to point to “studies” supporting their point of view, whether it’s on economic or cultural issues.  Or the real beating blackened heart of terrible science, diets.  The academy has many problems, but something like this would never make it past peer review because the design is obviously bunk to someone who understands the instrumental variable technique.  On the other hand, it can be released as a “working paper” and run around the world twice before the truth can put its proverbial boots on.

Lying with math is a hard problem to deal with.  The tools of social science can be used and abused, and often even a reasonable and well-intentioned researcher can create work that is willfully misinterpreted and abused by bad actors.  I wish I had some good answers as to what can be done.  Actual working academics have professional ethical responsibilities, and peer review to keep them honest.  Unfortunately, the people who are most likely to abuse the public trust are the ones sitting outside that system.  And misinformation is hard to correct once people have internalized it.

The Cuba Embargo and Public Opinion

Returning to blogging after a long hiatus and a pretty busy semester in which I actually got some research done!  Still recovering from the experience; it was exhausting, as it turns out.

As most newsreaders have heard by now, Obama will be normalizing relations with Cuba and attempting to end the fifty-year embargo.   On the substance, hard to see how it’s a bad move.  The embargo has gone on for long enough and the Castro regime seems as solid as ever, and it’s difficult to see what continuing the embargo for another decade or two could possibly accomplish.  More interesting to me are the domestic political consequences; everyone seems to have an opinion on whether or not this will destroy the Democratic chances to win Florida in 2016.  Two bigger points and one modest prediction.

First, reactions to events are poorly predicted by pre-event polling.  Most of the commentary tends to make reference to existing polling, which in this case shows an even split amongst Cuban-Americans on the question of whether or not the US should end the embargo.  The problem is that this is often a very poor predictor of whether or not polls taken next week – or in 2016 – will reflect a similar split amongst Cuban-Americans.  Very often, people’s opinions on something that sits at the back of their mind are very different than their reactions once it’s been leading the news for weeks and they have given it more thought. Why?

Because secondly, reactions to political decisions often depend as much on the context as on the decision itself.  Many people have well-developed opinions on issues, but most don’t – they make their opinions on questions like the Cuba embargo on the things said by the people they respect on these sorts of issues.  This is often called “cue-taking”; cue-givers can be media figures, social leaders, or partisan leaders.  Almost all issue opinions are colored by partisanship, and people generally line up behind the position taken by the elites of their party.

One upshot of this is that when one party is unified and the other is divided, the position of the unified party tends to carry the day with the populace in general.  That leads to the modest prediction: ending the embargo will likely be a fairly popular position and the Congress is unlikely to mount a successful push to reverse it. As long as Republican presidential candidates are publicly clashing over the issue, Republican voters are quite unlikely to solidify in opposition to it.  Most Democrats will line up behind the President on the question.  By 2016, this will no longer be particularly controversial and not a major election issue.

Of course, this could be completely wrong!  But division within the Republican party suggests the hardliners will lose the fight to make this a Republican priority.

Sponsored Content, Propaganda, & Public Behavior

Andrew Sullivan has been on a tirade recently against “sponsored content”.  His latest target – pro-Israel “advertorials” placed in Buzzfeed by ReThink Israel, an organization run by Sheldon Adelson.

It is generally assumed that propaganda is successful, but this relies on strong assumptions about media consumption and public opinion.  In order to actually affect behavior, a few links are necessary.  People must read or watch a given piece of media, people must process the media, people must remember the media, it must flow through in a non-trivial fashion to their expressed political beliefs, and finally it must actually change their actions.  In order to actually affect politics, the change in belief must be substantial enough for at least some people to alter their patterns of activism, donation, or voting.

This could be successful, but I’m pretty dubious.  Mr. Adelson spent a great deal of money (as much as $150M) on the 2012 election.  This was money spent in very short time period, on a single issue, with a very clear call to action – vote for Mitt Romney a few weeks from now.  He helped saturate the airwaves for a relatively small group of voters in swing states, again covering a very short time frame and with extremely strong messaging about how Barack Obama will destroy freedom & Israel.  There seems to be no noticeable change in the election directly attributable to Mr. Adelson’s spending.  In fact, as far as anybody can tell the money was basically wasted.  These milquetoast advertisements about how Israel is great to visit (accurate)

It’s not clear what the takeaway is for the Buzzfeed crowd.  On the one hand, given that it is so unlikely to have any sort of consequences on public opinion they can take the money with a clean conscience.  On the other, the fact that it has so little real weight makes it a more gratuitous insult to their journalistic ethics.  For those of you without much love for Adelson, you should probably feel relieved that he is blowing his money on such ineffectual things.

A Theory of Politics In a Single Graph

Andrew Sullivan highlights a fantastic little statistical anecdote: most people in American support gay marriage, and most of them think they’re in the minority of public opinion.  In contrast, most opponents of gay marriage think they’re in a majority – a mere 20% are aware that they are the minority.  Here are the number in each group who can correctly identify how many Americans support gay marriage:

This is a great little case study demonstrating the ambiguity of politics.  First and most obviously, people tend to believe they’re in the minority, especially when the cause is important to them.  Those who strongly favor gay marriage are the only ones right on this, but in a broken-clock-is-right-twice-a-day sense.  More generally, people believe what flatters them.  But the other thing it shows nicely is an explanation for “punctuated equilibrium” in politics.

This is an important question for students of politics – do things change gradually and continuously? Or are political transitions like bankruptcy; they happen  “slowly, and then all at once“?  This graph demonstrates why the second might be true.  Today gay marriage supporters are a majority, but conventional wisdom hasn’t yet incorporated that fact.  I’m guessing that the second change, which doesn’t rely on people changing deeply-held beliefs, will happen a lot faster.  And as it does we might well see a sudden sharp shift in outcomes as people realize where the median voter actually is.  We might see breathtakingly fast change in policy…like for example, a wave of judges across the country releasing a cascade of decisions finding an equal right to marriage.

A whole theory of political change from a single graph – isn’t social science neat?

Cyberwarfare and Public Accountability

One of the reasons that American presidents go to war – not a good reason, mind you – is known as the “Rally Around the Flag” effect.  When America gets involved in an armed conflict, whether as defender or aggressor, the president becomes more popular and more highly-approved, and often the conflict itself is accompanied by a burst of legislation unrelated to the war.  There’s a lot of debate over precisely how strong the effect is and what drives it, but the existence of the effect itself is one of the best-known findings of political science.

This could be relevant for the decision to embrace cyberwarfare.  Today the NYT reveals that the Obama adminstration was deeply divided over the question of whether to use cyberweapons to attack the Syrian government.  The NYT reveals the deep discussion over the strategic benefits and risks – but one that does not appear there is the potential effect on American public opinion.  Cyberwarfare is still visible to affected foreign players (and possibly friendly/neutral ones too) and America is strategically accountable for its actions in this sphere, but it can be plausibly denied in a way that bombers and paratroopers can’t.  If Obama had decided to go forward with attacks on Syria, he would have had to deal with the fallout from Syria and Russia, but it likely would have remained secret until the next Edward Snowden leaked it.

If cyberwarfare is normalized, more acts of national aggression will take place out of the public eye.  As a positive question – a question of facts – public opinion is a significant constraint on executive action.  As a normative question, people differ a lot on whether this constraint is a good or bad thing.  Perhaps the people stop wise Presidents from taking the actions necessary to protect the country; perhaps the people’s reluctance to go to war stops foolhardy Presidents from making dangerous leaps into conflict.

The growth of cyberwarfare will be a neat and potentially worrying test of who is right.

Are Values Stickier Than Prices?

Interestingly, Republicans seem to be more supportive of bombing Syria than Democrats, though both are on balance opposed.  This is unsurprising – Democrats are less supportive of bombing places in general but often line up behind the President, while Republicans are more supportive in general but don’t care for this Obama fellow.

But it brings up an interesting question – when and where does party consensus actually matter?  For example, in 2002-2008 the urgency of global warming was generally accepted by the Republican Party, leading to sights like the infamous Gingrich-Pelosi “couch” ad.  But upon Obama’s election, the opposition to any action has become completely hardened.  Or during the Bush Administration, Democrats were generally highly critical of expansive executive powers, while the Obama Administration has exploited the “imperial presidency” to its limits with barely a whisper of protest from Democrats.  However, other issues remain remarkably stable in partisan valence over decades – for example, the Democratic push for centralizing healthcare regulation and Republican opposition.

Obviously, different issues and belief sets have different values of “stickiness”.  Keynesians talk about “sticky” prices, which can’t be adjusted in response to changing economic conditions and ultimately drive firms into bankruptcy and economies into recession.  Well, some political values are similarly sticky, whereas others aren’t.  Republicans’ belief in lower taxes for the rich, and Democrats’ beliefs in higher taxes for the rich, are incredibly sticky.  Other beliefs like executive power, civil rights, and regulation are much less sticky and both the parties flip back and forth a lot – Nixon created the EPA and Clinton unleashed Wall Street, after all.

It could be worthwhile to look at how to measure the “inertia” of beliefs and values.  These seems like it should be possible, through historic public opinion polling and rates of change.  I would guess that the high-“salience” issues, things people care about a lot like taxes, are stable.  Other more peripheral issues like civil rights* bounce around a lot since peoples’ opinions are less well-developed.  Since they have lower inertia, they are much more responsive to changes in political leadership, convenience, and changes in circumstance.

More interestingly, do different parties over time have different total levels of inertia?  This could truly go either way, and I don’t have any real hypotheses.  It is certainly plausible that parties become much more flexible or less flexible over time based on their voting base composition.  Or perhaps parties in opposition are always more flexible.  Or perhaps their level of inertia is roughly constant – their core beliefs become that much more tightly held in response to tactical flexibility.  The Republican Party has flipped positions on many practical issues since Obama took office – but at the same time, Republicans seem to be taking much more seriously their core commitments to a low-tax, low-spending government.

Learning to recognize which value are sticky could be a boon to governance, as majorities could know what is worthwhile to bargain with minorities over instead of wasting time where no bargains are possible.

*: Describing them as “peripheral” is sad, but when was the last time you voted based solely on a civil rights concern?

“Small Business” Is Just as Bad (Or Good) As “Big Business”

It’s a commonly-cited fact that “small business” is generally more trusted in America than the government, academia, or even churches.  What is a small business that it is such a font of moral virtue?  According to the US, it is a business with fewer than 250 employees.  Apparently that is all it takes.

Not to Godwin’s Law myself, but most slave plantations were small businesses.  Large planters had over 50 slaves, but of course far fewer employees.  I wouldn’t describe Candyland as more morally praiseworthy than the Lowell Mills.  In that proud tradition , small farmers today are generally the ones responsible for the most egregious abuses of migrant labor.  An independent farm owner can get away with labor abuses that large corporations never could, because they aren’t closely monitored by OSHA.

Small business has all the same virtues and defects as larger businesses in smaller packages.  It is an interesting question as to how it has acquired such a patina of grace.

Truth in Opinion

Jonathan Bernstein drops some knowledge on debt ceiling polling.  Basically, pre-debt-ceiling-battle, respondents will agree with abstract statements about matching debt-ceiling hikes dollar-for-dollar with spending cuts.  This is not necessarily a good predictor of how people will react should Republicans force a debt default because Democrats refuse to cut Social Security.

This is a factor of a few things.  One is specificity – while “cutting spending” is popular, specific spending cuts are generally not.  The Republicans will not be able to hold together on the debt ceiling unless they go balls-to-the-walls on this –  hitting the Sunday shows, grandstanding in Congress, endless CNN appearances, all threatening national default unless Democrats consent to their demands on cutting Social Security.  And they won’t be able to play the game of demanding Obama specify the cuts, because he has declined to do so and would be dumb to cede that ground.  In order to get their base appropriately riled up, they need to make demands and fight for them.  Perhaps Republicans think the terms of that fight favor them, but I doubt that.

More fundamentally, people are really bad at judging their future behavior.  I’m not an authority, but I did write my undergrad thesis on public opinion dynamics and I now do market research for my job. If there’s anything I’ve learned, it’s that people are really crappy at predicting their future opinions and behavior.  Asking people today, with relatively leading wording, how they would react should the Republicans cause a debt default is a terrible predictor of how people would react should the Republicans cause a debt default.

Hopefully John Boehner is dropping surveys like that to fire up the troops and intimidate the Democrats, rather than as an expression of his actual beliefs.  Given his aversion from maximalism in confrontations before now, I’m guessing the former and I hope that I’m right.  Making overconfident predictions about a basically unthinkable action is the sort of thing that gets one into trouble.